With gradual popularization of large-screen and super-screen intelligent terminals, a processor of an intelligent terminal has an increasingly high processing capability, so that many applications, for implementing a control based on man-machine interaction, appear on a large screen or a super large screen. In a process of implementing a control based on man-machine interaction, multiple users may run different interaction modes, by creating groups in one-to-one, one-to-many, and many-to-many forms, to obtain different interaction results. For example, in a graphical user interface obtained through rendering on a large screen or a super large screen, after multiple users are grouped into two different groups, by controlling the processing in man-machine interaction, information exchange may be performed between different groups, and different interaction results are obtained according to a response to information exchange; and by controlling the processing in man-machine interaction, information exchange may further be performed between group members in a same group, and different interaction results are obtained according to a response to information exchange.
Conventionally, release of a particular skill may be triggered to enrich a presentation form and content of information, and different presentation forms and content of information may finally lead to different interaction results. However, when releasing a particular skill, a skill release location and direction cannot be accurately determined, leading to occurrence of a large quantity of mis-operations, and affecting precision and accuracy of interaction processing.